Surround view camera calibration method, system, device, vehicle and storage medium

Abstract

The present invention discloses a surround-view camera calibration method, system, device, vehicle and storage medium, and belongs to the field of vehicle assisted driving technology. The present invention obtains the collected images collected by all surround-view cameras, splices the collected images to obtain a spliced ​​image, and then performs image recognition on the spliced ​​image to obtain a recognition image for the spliced ​​image, so as to match the recognition image with a preset image standard to obtain an image matching result, and determines whether the image quality collected by the vehicle surround-view camera meets the preset standard according to the matching with the preset image standard, and then determines the number of target corner points based on the image matching result, and determines the number of target corner points on the basis of ensuring the image quality of the collected image, so as to calibrate the surround-view camera according to the number of target corner points. Under the premise of meeting the preset image standard, the number of target corner points is determined to improve the production line pass rate, thereby improving the production efficiency of the vehicle.

Description

Technical Field

[0001] The present invention relates to the field of vehicle assisted driving technology, and in particular to a surround view camera calibration method, system, device, vehicle and storage medium. Background Art

[0002] With the development of science and technology, people have put forward higher requirements for smart cars. In order to avoid the impact of blind spots on vehicle driving safety, the visual assistance solution of installing multiple cameras around the car body to form a panoramic view system has been applied to many models. As an important component of the parking assistance system, the on-board panoramic view system is to display the full picture of the car's surrounding environment in real time, provide the driver with dynamic images around the car body without blind spots, and improve driving safety. The on-board panoramic view system is generally composed of four cameras. The images collected by the four cameras are spliced ​​and processed to form an image of the vehicle's surrounding environment. In the production process of models equipped with on-board panoramic view systems, the four cameras need to be calibrated to ensure the image splicing effect. At present, when calibrating the camera, the black grid points on the calibration plate around the car are mainly used, but the camera is usually far away from the black grid points. When recognizing the image collected by the camera, the corner points cannot be recognized. When the number of corner points detected does not meet the requirements, it will be considered that the camera calibration has failed, resulting in a low pass rate of the production line and affecting the vehicle production efficiency. Summary of the invention

[0003] The main purpose of the present invention is to provide a surround view camera calibration method, system, device, vehicle and storage medium, aiming to solve the technical problem of how to improve vehicle production efficiency.

[0004] To achieve the above object, the present invention provides a surround view camera calibration method, the surround view camera calibration method comprising the following steps:

[0005] Acquire the collected images collected by all surround view cameras, stitch the collected images together to obtain a stitched image, and perform image recognition on the stitched image to obtain a recognized image;

[0006] Matching the identified image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result;

[0007] The surround view camera is calibrated according to the number of target corner points.

[0008] Optionally, before the step of stitching the collected images to obtain a stitched image, the step further includes:

[0009] Determining whether each of the captured images contains a preset grid area;

[0010] If the preset grid area is included, the step of splicing the collected images to obtain a spliced ​​image is performed.

[0011] Optionally, the step of matching the recognized image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result includes:

[0012] determining whether the identified image matches a preset image standard;

[0013] If the recognition image matches the preset image standard, determining a first preset number of corner points corresponding to the recognition image, reducing the first preset number of corner points by a preset number to obtain a second preset number of corner points;

[0014] Performing secondary recognition on the stitched image based on the second preset number of corner points to obtain an updated recognition image, and determining whether the updated recognition image matches the preset image standard;

[0015] If the updated recognition image does not match the preset image standard, the first preset number of corner points is used as the target number of corner points.

[0016] Optionally, the step of determining whether the recognized image matches a preset image standard comprises:

[0017] Determining the number of first corner points identified in the identified image;

[0018] Whether the recognized image matches a preset image standard is determined based on whether the first recognized corner point quantity matches the first preset corner point quantity.

[0019] Optionally, the step of acquiring the collected images collected by all surround view cameras and stitching the collected images to obtain a stitched image includes:

[0020] Controlling the vehicle to enter a preset calibration area, and using an image corresponding to the preset calibration area among the images captured by the surround view camera as a captured image;

[0021] The captured images are stretched and cropped according to a preset splicing rule, and the stretched and cropped captured images are spliced ​​to generate a spliced ​​image.

[0022] In addition, to achieve the above-mentioned purpose, the present invention further provides a surround view camera calibration system, the surround view camera calibration system comprising a grid calibration plate, a vehicle, a surround view camera and an image recognition device,

[0023] The grid calibration plate is arranged around the vehicle;

[0024] The surround view camera is connected to the vehicle and is used to collect images around the vehicle;

[0025] The image recognition device is connected to the surround-view camera and is used to perform image recognition on the images collected by the surround-view camera.

[0026] Optionally, the surround-view cameras include four, which are respectively located at the front grille of the vehicle, the left exterior rearview mirror of the vehicle, the right exterior rearview mirror of the vehicle, and the rear tailgate of the vehicle.

[0027] In addition, to achieve the above-mentioned purpose, the present invention further provides a surround view camera calibration device, the surround view camera calibration device comprising:

[0028] An image acquisition module is used to acquire images acquired by all surround view cameras, stitch the acquired images together to obtain a stitched image, and perform image recognition on the stitched image to obtain a recognized image;

[0029] A corner point determination module, used for matching the recognition image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result;

[0030] The calibration execution module is used to calibrate the surround view camera according to the number of target corner points.

[0031] In addition, to achieve the above-mentioned purpose, the present invention also provides a vehicle, comprising a memory, a processor, and a surround-view camera calibration program stored in the memory and executable on the processor, wherein: when the surround-view camera calibration program is executed by the processor, the steps of the surround-view camera calibration method as described above are implemented.

[0032] In addition, to achieve the above-mentioned purpose, the present invention also provides a storage medium, on which a surround view camera calibration program is stored, and when the surround view camera calibration program is executed by a processor, the steps of the surround view camera calibration method as described above are implemented.

[0033] The present invention proposes a surround-view camera calibration method, system, device, vehicle and storage medium, which obtains the collected images collected by all surround-view cameras, splices the collected images to obtain a spliced ​​image, and provides an image source for image recognition, so as to perform image recognition on the spliced ​​image, obtain a recognition image for the spliced ​​image, match the recognition image with a preset image standard, obtain an image matching result, and determine whether the image quality collected by the vehicle surround-view camera meets the preset standard based on the matching with the preset image standard, and then determine the number of target corner points based on the image matching result, and determine the number of target corner points on the basis of ensuring the image quality of the collected image, so as to calibrate the surround-view camera based on the number of target corner points. Under the premise of meeting the preset image standard, the number of target corner points is determined to improve the production line pass rate, thereby improving the production efficiency of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the structure of a vehicle in a hardware operating environment involved in an embodiment of the present invention;

[0035] Figure 2 It is a schematic diagram of the flow chart of the first embodiment of the surround view camera calibration method of the present invention;

[0036] Figure 3 It is a schematic diagram of the application scenario of the present invention;

[0037] Figure 4 Schematic diagram of the structure of the surround view camera calibration system of the present invention;

[0038] Figure 5 It is a schematic diagram of the structure of the surround view camera calibration device of the present invention.

[0039] Description of Figure Numbers:

[0040]

[0041] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0042] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0043] Reference Figure 1 , Figure 1 A schematic diagram of the vehicle structure of the hardware operating environment involved in the embodiment of the present invention.

[0044] like Figure 1As shown, the vehicle may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0045] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the vehicle, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0046] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a data storage module, a network communication module, a user interface module, and a surround view camera calibration program.

[0047] exist Figure 1 In the vehicle shown, the network interface 1004 is mainly used for data communication with other devices; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the vehicle of the present invention can be set in the vehicle, and the vehicle calls the surround-view camera calibration program stored in the memory 1005 through the processor 1001, and executes the surround-view camera calibration method provided in an embodiment of the present invention.

[0048] The embodiment of the present invention provides a surround view camera calibration method, referring to Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of a surround view camera calibration method according to the present invention.

[0049] In this embodiment, the surround view camera calibration method includes:

[0050] Step S100, acquiring images acquired by all surround view cameras, stitching the acquired images to obtain a stitched image, and performing image recognition on the stitched image to obtain a recognized image;

[0051] In this embodiment, it should be noted that the panoramic surround view system sets up four to eight wide-angle cameras around the car that can cover all the fields of view around the vehicle, processes the multi-channel video images collected at the same time into a 360-degree overhead view of the vehicle, and finally displays it on the screen of the center console, so that the driver can clearly check whether there are obstacles around the vehicle and understand the relative position and distance of the obstacles, helping the driver to park the vehicle easily. Not only is it very intuitive, but there are no blind spots, which can improve the driver's calm control of the vehicle to park or pass through complex roads, and effectively reduce the occurrence of accidents such as scratches, collisions, and sinking. The panoramic surround view system may include multiple cameras for collecting images around the vehicle. In this embodiment, the panoramic surround view system includes four cameras as an example for explanation. The four cameras respectively capture images of the front, back, left, and right of the vehicle, and generate panoramic image information of the car and its surrounding environment after processing the collected images. In the vehicle production process, the cameras in the panoramic surround view system need to be calibrated to ensure the accuracy of image acquisition. The surround view camera refers to the camera in the panoramic surround view system; the collected image refers to the image captured by the camera. When calibrating the surround view camera, it is necessary to predetermine the corner point recognition standard used to ensure the quality of image stitching during calibration, that is, determine the number of corner points in the corner point recognition standard, and calibrate the surround view camera according to the corner point recognition standard. Specifically, obtain the collected images collected by all the surround view cameras in the panoramic surround view system, and stitch the collected images to obtain a stitched image; perform image recognition on the stitched image, and generate a recognition image based on the recognition result.

[0052] Step S200, matching the identified image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result;

[0053] In this embodiment, it should be noted that the preset image standard refers to the minimum recognition standard for recognizing that the image quality of the current image is good. The preset image standard is determined based on a large number of experiments and can be changed according to actual needs. The number of target corner points refers to the number of corner points in the corner point recognition standard. The number of target corner points is determined based on the matching result of matching the recognition image with the preset image standard. Among them, the image matching result is that the recognition image matches the preset image standard, or the recognition image does not match the preset image standard. Determining the number of target corner points based on the image matching result is to minimize the number of corner points while ensuring image quality, so as to improve production efficiency. Specifically, after generating the recognition image, the recognition image is matched with the preset image standard to determine the image matching result, and the number of target corner points is determined based on the image matching result.

[0054] Step S300, calibrating the surround view camera according to the number of target corner points.

[0055] In this embodiment, it should be noted that the calibration of the surround view camera is a process of calibrating the surround view camera. Figure 3 , by setting a grid calibration plate at the calibration station of the vehicle production line, determining the calibration area, placing the vehicle to be calibrated with the on-board panoramic surround view system in the calibration area, so that the grid calibration plate is located around the vehicle to be calibrated. The grid calibration plate is provided with square black grid points, and the black grid points in the image captured by the camera are subjected to corner point recognition to determine whether the number of identified corner points meets the target number of corner points. If the number of corner points recognized for the black grid points meets the target number of corner points, the calibration of the surround view camera of the vehicle is considered to be completed. If the number of corner points recognized does not meet the target number of corner points, the surround view camera of the vehicle needs to be further debugged until the number of corner points recognized meets the target number of corner points. Specifically, after determining the target number of corner points, the surround view camera is calibrated according to the target number of corner points.

[0056] In an embodiment of the present invention, the collected images collected by all surround-view cameras are obtained, and the collected images are spliced ​​to obtain a spliced ​​image to provide an image source for image recognition, so as to perform image recognition on the spliced ​​image to obtain a recognition image for the spliced ​​image, and match the recognition image with a preset image standard to obtain an image matching result, and determine whether the image quality collected by the vehicle surround-view camera meets the preset standard based on the matching with the preset image standard, and then determine the number of target corner points based on the image matching result, and determine the number of target corner points on the basis of ensuring the image quality of the collected image, so as to calibrate the surround-view camera based on the number of target corner points. Under the premise of meeting the preset image standard, the number of target corner points is determined to improve the production line pass rate, thereby improving the production efficiency of the vehicle.

[0057] Furthermore, based on the first embodiment of the surround view camera calibration method of the present invention, a second embodiment of the surround view camera calibration method of the present invention is proposed, and before the step of stitching the collected images to obtain the stitched image, the step further includes:

[0058] Step a, determining whether each of the collected images contains a preset grid area;

[0059] Step b: If the preset grid area is included, executing the step of splicing the collected images to obtain a spliced ​​image.

[0060] In this embodiment, it should be noted that when calibrating the surround view cameras of the panoramic surround view system of the vehicle, the vehicle is located at the calibration station, that is, there are grid calibration plates around the vehicle. In order to cover the field of view around the vehicle as much as possible, the panoramic surround view system includes multiple cameras, and the multiple cameras are respectively located at multiple positions of the vehicle. In this embodiment, four surround view cameras are used as an example, and the surround view cameras are respectively set at the front, back, left and right positions of the vehicle to collect image data around the vehicle. Since the cameras respectively collect images relative to different positions of the vehicle, it is necessary to splice the images collected by each camera to obtain a spliced ​​image including all image data around the vehicle. The surround view cameras are respectively located at different positions of the vehicle. In order to ensure the integrity of the spliced ​​image, it is necessary to ensure that the intersection area of ​​the adjacent surround view camera capture images is identified. The preset grid area refers to the intersection area between adjacent cameras. By ensuring that the captured image of each surround view camera contains the preset grid area, the integrity of the spliced ​​image spliced ​​based on each captured image is ensured.

[0061] Specifically, determine whether the captured images captured by each surround-view camera contain a preset grid area. If so, perform a step of stitching the captured images to obtain a stitched image. If not, debug the surround-view cameras that do not have a preset grid area in the captured images to ensure that each surround-view camera has a preset grid area in the captured images.

[0062] In this embodiment, the integrity of the captured image is ensured and the quality of the spliced ​​image is improved by determining the preset grid area.

[0063] Furthermore, the steps of matching the identified image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result include:

[0064] Step c, determining whether the identified image matches a preset image standard;

[0065] Step d, if the recognition image matches the preset image standard, determining a first preset number of corner points corresponding to the recognition image, reducing the first preset number of corner points by a preset number to obtain a second preset number of corner points;

[0066] Step e, performing secondary recognition on the stitched image based on the second preset number of corner points to obtain an updated recognition image, and determining whether the updated recognition image matches the preset image standard;

[0067] Step f: If the updated recognition image does not match the preset image standard, the first preset number of corner points is used as the target number of corner points.

[0068] In this embodiment, it should be noted that the preset number of corner points refers to the corner point recognition standard used to ensure the quality of image stitching when calibrating the surround view camera. The first preset number of corner points is the corner point recognition standard set during the calibration process of the current surround view camera. When recognizing an image, the quality of the recognized image can be regulated based on the preset number of corner points. When recognizing an image, the more preset corner points there are, the higher the requirements for the recognized image. In this embodiment, under the premise of ensuring the quality of the recognized image, the preset number of corner points is adjusted to the maximum extent, so that the recognized image is limited by the minimum number of preset corner points, thereby improving the production line pass rate. The preset image standard refers to the minimum recognition standard for determining that the image quality of the current image is good. The preset image standard is determined based on a large number of experiments and can be changed according to actual needs. The preset number refers to the preset value for reducing the number of preset corner points at a time. The second preset number of corner points is the number of corner points reduced by the preset number based on the first preset number of corner points. The updated recognized image refers to the image obtained after image recognition under the restriction of the second preset number of corner points.

[0069] Specifically, after acquiring the captured images captured by the camera, the captured images are spliced, and image recognition is performed under the restriction of the first preset number of corner points to obtain a recognized image. The recognized image is matched with the preset image standard, and the matching rule is whether the recognized image meets the preset image standard. If the recognized image matches the preset image standard, it can be considered that the current recognized image has good quality, and the first preset number of corner points to be restricted when performing the image recognition is determined, and the first preset number of corner points is reduced to reduce the preset number to obtain the second preset number of corner points. The spliced ​​image is recognized twice according to the restriction of the second preset number of corner points to obtain an updated recognized image, and it is determined whether the updated recognized image matches the preset image standard. If the updated recognized image does not match the preset image standard, that is, the quality of the updated recognized image does not meet the preset image standard, the number of corner points corresponding to the second preset number of corner points is too small. In order to ensure the quality of the recognized image, the first preset number of corner points before the number of corner points is reduced is used as the target number of corner points.

[0070] It is understandable that when the updated recognition image is matched with the preset image standard, if the updated recognition image matches with the preset image standard, the preset number can be further reduced based on the second preset number of corner points to obtain a third preset number of corner points, and the spliced ​​image is further image recognized based on the third preset number of corner points to obtain three recognition images, and the three recognition images are matched with the preset image standard, and so on, the target number of corner points is determined based on the matching result. The target number of corner points is the last preset number of corner points before the preset number is reduced from the preset number of corner points corresponding to the recognition image that does not match the preset image standard.

[0071] In this embodiment, the number of target corner points is determined by matching the recognition image with the preset image standard, thereby minimizing the number of corner points while ensuring the quality of the recognition image, increasing the pass rate of production line camera calibration, and improving production efficiency.

[0072] Furthermore, the step of determining whether the recognized image matches a preset image standard includes:

[0073] Step g, determining the number of first corner points identified in the identified image;

[0074] Step h: determining whether the recognized image matches a preset image standard based on whether the first recognized corner point quantity matches the first preset corner point quantity.

[0075] In this embodiment, it should be noted that the first corner point recognition number refers to the number of corner points recognized after the recognition image is recognized. Whether the first corner point recognition number matches the first preset corner point number refers to whether the first corner point recognition number reaches the number value corresponding to the first preset corner point number. If the first corner point recognition number meets the first preset corner point number, it is determined that the recognition image matches the preset image standard; if the first corner point recognition number does not meet the first preset corner point number, it is determined that the recognition image does not match the preset image standard. The quality of the recognition image is ensured by the number of corner point recognitions, and the preset number of corner point recognitions is adjustable, which improves the flexibility of image quality determination.

[0076] Furthermore, the steps of acquiring the collected images collected by all the surround view cameras and stitching the collected images to obtain the stitched image include:

[0077] Step i, controlling the vehicle to enter a preset calibration area, and taking an image corresponding to the preset calibration area among the images captured by the surround view camera as a captured image;

[0078] Step j: stretching and cropping the captured images according to a preset stitching rule, and stitching the stretched and cropped captured images to generate a stitched image.

[0079] In this embodiment, it should be noted that by setting a grid calibration plate at the calibration station of the vehicle production line, a preset calibration area is determined, and the vehicle to be calibrated equipped with a vehicle-mounted panoramic surround view system is controlled to drive into the preset calibration area, so that the grid calibration plate is located around the vehicle to be calibrated. At this time, the surround view camera on the vehicle performs image acquisition. In order to reduce the errors that may be caused by other variable factors, the images related to the preset calibration area in the acquired images are used as acquired images. The preset stitching rules refer to the standards for image stitching, and the preset stitching rules can be changed according to actual needs. The acquired images are stretched and cropped according to the preset stitching rules, and the stretched and cropped acquired images are stitched to generate a stitched image.

[0080] In this embodiment, the captured images are restricted to reduce irrelevant variables, thereby ensuring the accuracy of image capture, and the captured images are processed and then spliced ​​to improve the visibility and effectiveness of the spliced ​​images.

[0081] In addition, reference Figure 4 The present invention also proposes a surround camera calibration system 01, which includes a grid calibration plate 010, a vehicle 020, a surround camera 030, and an image recognition device 040. The grid calibration plate 010 is arranged around the vehicle 020, and the surround camera 030 is connected to the vehicle 020. The surround camera 030 includes four surround cameras, which are respectively located at the front grille position of the vehicle 020, the left exterior rearview mirror position of the vehicle 020, the right exterior rearview mirror position of the vehicle 020, and the rear tailgate position of the vehicle 020. The image recognition device 040 is connected to the surround camera 030. The surround camera 030 collects images of the grid calibration plate around the vehicle 020 and sends the image to the image recognition device 040; the image recognition device 040 performs image recognition on the image collected by the surround camera 030. Among them, the position of the surround camera 030 can also be adjusted according to different vehicle models and actual needs. In one embodiment, the surround camera calibration method of the present invention is applied to the surround camera calibration system of the present invention. It is understandable that, in one embodiment, the surround view camera calibration system 01 may also include more or fewer components.

[0082] In addition, reference Figure 5 The present invention also proposes a surround view camera calibration device, the device comprising:

[0083] The image acquisition module 2001 is used to acquire the collected images collected by all the surround view cameras, stitch the collected images together to obtain a stitched image, and perform image recognition on the stitched image to obtain a recognized image;

[0084] A corner point determination module 2002 is used to match the recognition image with a preset image standard to obtain an image matching result, and determine the number of target corner points based on the image matching result;

[0085] The calibration execution module 2003 is used to calibrate the surround view camera according to the number of target corner points.

[0086] Optionally, the image acquisition module 2001 is further used for:

[0087] Determining whether each of the captured images contains a preset grid area;

[0088] If the preset grid area is included, the step of splicing the collected images to obtain a spliced ​​image is performed.

[0089] Optionally, the image acquisition module 2001 is further used for:

[0090] Controlling the vehicle to enter a preset calibration area, and using an image corresponding to the preset calibration area among the images captured by the surround view camera as a captured image;

[0091] The captured images are stretched and cropped according to a preset splicing rule, and the stretched and cropped captured images are spliced ​​to generate a spliced ​​image.

[0092] Optionally, the corner point determination module 2002 is further configured to:

[0093] determining whether the identified image matches a preset image standard;

[0094] If the recognition image matches the preset image standard, determining a first preset number of corner points corresponding to the recognition image, reducing the first preset number of corner points by a preset number to obtain a second preset number of corner points;

[0095] Performing secondary recognition on the stitched image based on the second preset number of corner points to obtain an updated recognition image, and determining whether the updated recognition image matches the preset image standard;

[0096] If the updated recognition image does not match the preset image standard, the first preset number of corner points is used as the target number of corner points.

[0097] Optionally, the corner point determination module 2002 is further configured to:

[0098] Determining the number of first corner points identified in the identified image;

[0099] Whether the recognized image matches a preset image standard is determined based on whether the first recognized corner point quantity matches the first preset corner point quantity.

[0100] The specific implementation of the surround view camera calibration device of the present invention is basically the same as the above-mentioned embodiments of the surround view camera calibration method, and will not be repeated here.

[0101] In addition, the present invention also proposes a vehicle, characterized in that the vehicle includes a memory, a processor, and a surround-view camera calibration program stored in the memory and executable on the processor, wherein: when the surround-view camera calibration program is executed by the processor, the surround-view camera calibration method described in each embodiment of the present invention is implemented.

[0102] In addition, the present invention also provides a storage medium on which a surround view camera calibration program is stored. The storage medium may be Figure 1 The memory 20 in the terminal may also be at least one of a ROM (Read-Only Memory) / RAM (Random Access Memory), a magnetic disk, and an optical disk, and the storage medium includes a plurality of instructions for enabling a vehicle with a processor to execute the surround view camera calibration method described in each embodiment of the present invention.

[0103] It is understood that, in the description of this specification, the description with reference to the terms "one embodiment", "another embodiment", "other embodiments", or "first to Nth embodiments" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0104] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0105] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0106] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0107] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A surround view camera calibration method, characterized in that: The surround view camera calibration method comprises the following steps: Acquire the collected images collected by all surround view cameras, stitch the collected images together to obtain a stitched image, and perform image recognition on the stitched image to obtain a recognized image; Matching the identified image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result; Calibrate the surround camera according to the number of target corner points; The step of matching the identified image with a preset image standard to obtain an image matching result, and determining the number of target corner points based on the image matching result comprises: determining whether the identified image matches a preset image standard; If the recognition image matches the preset image standard, determining a first preset number of corner points corresponding to the recognition image, reducing the first preset number of corner points by a preset number to obtain a second preset number of corner points; Performing secondary recognition on the stitched image based on the second preset number of corner points to obtain an updated recognition image, and determining whether the updated recognition image matches the preset image standard; If the updated recognition image does not match the preset image standard, the first preset number of corner points is used as the target number of corner points.

2. The surround view camera calibration method according to claim 1, characterized in that: Before the step of stitching the collected images to obtain a stitched image, the method further includes: Determining whether each of the captured images contains a preset grid area; If the preset grid area is included, the step of splicing the collected images to obtain a spliced ​​image is performed.

3. The surround view camera calibration method according to claim 1, characterized in that: The step of determining whether the identified image matches a preset image standard comprises: Determining the number of first corner points identified in the identified image; Whether the recognized image matches a preset image standard is determined based on whether the first recognized corner point quantity matches the first preset corner point quantity.

4. The surround view camera calibration method according to claim 1, characterized in that: The step of acquiring the collected images collected by all the surround view cameras and stitching the collected images to obtain a stitched image comprises: Controlling the vehicle to enter a preset calibration area, and using an image corresponding to the preset calibration area among the images captured by the surround view camera as a captured image; The captured images are stretched and cropped according to a preset splicing rule, and the stretched and cropped captured images are spliced ​​to generate a spliced ​​image.

5. A surround view camera calibration system, characterized in that: The surround view camera calibration system can implement the steps of the surround view camera calibration method according to any one of claims 1 to 4, and the surround view camera calibration system includes a grid calibration plate, a vehicle, a surround view camera, and an image recognition device. The grid calibration plate is arranged around the vehicle; The surround view camera is connected to the vehicle and is used to collect images around the vehicle; The image recognition device is connected to the surround-view camera and is used to perform image recognition on the images collected by the surround-view camera.

6. The surround view camera calibration system according to claim 5, characterized in that: The surround view cameras include four, which are respectively located at the front grille of the vehicle, the left exterior rearview mirror of the vehicle, the right exterior rearview mirror of the vehicle, and the rear tailgate of the vehicle.

7. A surround view camera calibration device, characterized in that: The surround view camera calibration device comprises: An image acquisition module is used to acquire images acquired by all surround view cameras, stitch the acquired images together to obtain a stitched image, and perform image recognition on the stitched image to obtain a recognized image; A corner point determination module is used to match the recognition image with a preset image standard to obtain an image matching result, and determine the target corner point number based on the image matching result; the corner point determination module is also used to determine whether the recognition image matches the preset image standard; if the recognition image matches the preset image standard, determine a first preset corner point number corresponding to the recognition image, reduce the first preset corner point number by a preset number, and obtain a second preset corner point number; perform secondary recognition on the spliced ​​image based on the second preset corner point number to obtain an updated recognition image, and determine whether the updated recognition image matches the preset image standard; if the updated recognition image does not match the preset image standard, use the first preset corner point number as the target corner point number; The calibration execution module is used to calibrate the surround view camera according to the number of target corner points.

8. A vehicle, characterized in that: The vehicle comprises: a memory, a processor, and a surround view camera calibration program stored in the memory and executable on the processor, wherein the surround view camera calibration program is configured to implement the steps of the surround view camera calibration method according to any one of claims 1 to 4.

9. A storage medium, characterized in that: The storage medium stores a surround view camera calibration program, and when the surround view camera calibration program is executed by the processor, the steps of the surround view camera calibration method according to any one of claims 1 to 4 are implemented.

Images